yt_dlp/extractor/odnoklassniki.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_etree_fromstring,
   6     compat_parse_qs,
   7     compat_urllib_parse_unquote,
   8     compat_urllib_parse_urlparse,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     HEADRequest,
  13     float_or_none,
  14     int_or_none,
  15     qualities,
  16     smuggle_url,
  17     traverse_obj,
  18     unescapeHTML,
  19     unified_strdate,
  20     unsmuggle_url,
  21     url_or_none,
  22     urlencode_postdata,
  23 )
  24
  25
  26 class OdnoklassnikiIE(InfoExtractor):
  27     _VALID_URL = r'''(?x)
  28                 https?://
  29                     (?:(?:www|m|mobile)\.)?
  30                     (?:odnoklassniki|ok)\.ru/
  31                     (?:
  32                         video(?P<embed>embed)?/|
  33                         web-api/video/moviePlayer/|
  34                         live/|
  35                         dk\?.*?st\.mvId=
  36                     )
  37                     (?P<id>[\d-]+)
  38                 '''
  39     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1']
  40     _TESTS = [{
  41         'note': 'Coub embedded',
  42         'url': 'http://ok.ru/video/1484130554189',
  43         'info_dict': {
  44             'id': '1keok9',
  45             'ext': 'mp4',
  46             'timestamp': 1545580896,
  47             'view_count': int,
  48             'thumbnail': r're:^https?://.*\.jpg$',
  49             'title': 'Народная забава',
  50             'uploader': 'Nevata',
  51             'upload_date': '20181223',
  52             'age_limit': 0,
  53             'uploader_id': 'nevata.s',
  54             'like_count': int,
  55             'duration': 8.08,
  56             'repost_count': int,
  57         },
  58     }, {
  59         'note': 'vk.com embedded',
  60         'url': 'https://ok.ru/video/3568183087575',
  61         'info_dict': {
  62             'id': '-165101755_456243749',
  63             'ext': 'mp4',
  64             'uploader_id': '-165101755',
  65             'duration': 132,
  66             'timestamp': 1642869935,
  67             'upload_date': '20220122',
  68             'thumbnail': str,
  69             'title': str,
  70             'uploader': str,
  71         },
  72         'skip': 'vk extractor error',
  73     }, {
  74         # metadata in JSON, webm_dash with Firefox UA
  75         'url': 'http://ok.ru/video/20079905452',
  76         'md5': '8f477d8931c531374a3e36daec617b2c',
  77         'info_dict': {
  78             'id': '20079905452',
  79             'ext': 'webm',
  80             'title': 'Культура меняет нас (прекрасный ролик!))',
  81             'thumbnail': str,
  82             'duration': 100,
  83             'upload_date': '20141207',
  84             'uploader_id': '330537914540',
  85             'uploader': 'Виталий Добровольский',
  86             'like_count': int,
  87             'age_limit': 0,
  88         },
  89         'params': {
  90             'format': 'bv[ext=webm]',
  91             'http_headers': {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; rv:102.0) Gecko/20100101 Firefox/102.0'},
  92         },
  93     }, {
  94         # metadataUrl
  95         'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
  96         'md5': '2bae2f58eefe1b3d26f3926c4a64d2f3',
  97         'info_dict': {
  98             'id': '63567059965189-0',
  99             'ext': 'mp4',
 100             'title': 'Девушка без комплексов ...',
 101             'thumbnail': str,
 102             'duration': 191,
 103             'upload_date': '20150518',
 104             'uploader_id': '534380003155',
 105             'uploader': '☭ Андрей Мещанинов ☭',
 106             'like_count': int,
 107             'age_limit': 0,
 108             'start_time': 5,
 109         },
 110         'params': {'skip_download': 'm3u8'},
 111     }, {
 112         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
 113         'url': 'https://ok.ru/video/3952212382174',
 114         'md5': '5fb5f83ce16cb212d6bf887282b5da53',
 115         'info_dict': {
 116             'id': '5axVgHHDBvU',
 117             'ext': 'mp4',
 118             'title': 'Youtube-dl 101: What is it and HOW to use it! Full Download Walkthrough and Guide',
 119             'description': 'md5:b57209eeb9d5c2f20c984dfb58862097',
 120             'uploader': 'Lod Mer',
 121             'uploader_id': '575186401502',
 122             'duration': 1529,
 123             'age_limit': 0,
 124             'upload_date': '20210405',
 125             'comment_count': int,
 126             'live_status': 'not_live',
 127             'view_count': int,
 128             'thumbnail': 'https://i.mycdn.me/i?r=AEHujHvw2RjEbemUCNEorZbxYpb_p_9AcN2FmGik64Krkcmz37YtlY093oAM5-HIEAt7Zi9s0CiBOSDmbngC-I-k&fn=external_8',
 129             'uploader_url': 'https://www.youtube.com/@MrKewlkid94',
 130             'channel_follower_count': int,
 131             'tags': ['youtube-dl', 'youtube playlists', 'download videos', 'download audio'],
 132             'channel_id': 'UCVGtvURtEURYHtJFUegdSug',
 133             'like_count': int,
 134             'availability': 'public',
 135             'channel_url': 'https://www.youtube.com/channel/UCVGtvURtEURYHtJFUegdSug',
 136             'categories': ['Education'],
 137             'playable_in_embed': True,
 138             'channel': 'BornToReact',
 139         },
 140     }, {
 141         # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
 142         'url': 'http://ok.ru/video/62036049272859-0',
 143         'info_dict': {
 144             'id': '62036049272859-0',
 145             'ext': 'mp4',
 146             'title': 'МУЗЫКА     ДОЖДЯ .',
 147             'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
 148             'upload_date': '20120106',
 149             'uploader_id': '473534735899',
 150             'uploader': 'МARINA D',
 151             'age_limit': 0,
 152         },
 153         'params': {
 154             'skip_download': True,
 155         },
 156         'skip': 'Video has not been found',
 157     }, {
 158         'note': 'Only available in mobile webpage',
 159         'url': 'https://m.ok.ru/video/2361249957145',
 160         'info_dict': {
 161             'id': '2361249957145',
 162             'ext': 'mp4',
 163             'title': 'Быковское крещение',
 164             'duration': 3038.181,
 165             'thumbnail': r're:^https?://i\.mycdn\.me/videoPreview\?.+',
 166         },
 167     }, {
 168         'note': 'subtitles',
 169         'url': 'https://ok.ru/video/4249587550747',
 170         'info_dict': {
 171             'id': '4249587550747',
 172             'ext': 'mp4',
 173             'title': 'Small Country An African Childhood (2020) (1080p) +subtitle',
 174             'uploader': 'Sunflower Movies',
 175             'uploader_id': '595802161179',
 176             'upload_date': '20220816',
 177             'duration': 6728,
 178             'age_limit': 0,
 179             'thumbnail': r're:^https?://i\.mycdn\.me/videoPreview\?.+',
 180             'like_count': int,
 181             'subtitles': dict,
 182         },
 183         'params': {
 184             'skip_download': True,
 185         },
 186     }, {
 187         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
 188         'only_matching': True,
 189     }, {
 190         'url': 'http://www.ok.ru/video/20648036891',
 191         'only_matching': True,
 192     }, {
 193         'url': 'http://www.ok.ru/videoembed/20648036891',
 194         'only_matching': True,
 195     }, {
 196         'url': 'http://m.ok.ru/video/20079905452',
 197         'only_matching': True,
 198     }, {
 199         'url': 'http://mobile.ok.ru/video/20079905452',
 200         'only_matching': True,
 201     }, {
 202         'url': 'https://www.ok.ru/live/484531969818',
 203         'only_matching': True,
 204     }, {
 205         'url': 'https://m.ok.ru/dk?st.cmd=movieLayer&st.discId=863789452017&st.retLoc=friend&st.rtu=%2Fdk%3Fst.cmd%3DfriendMovies%26st.mode%3Down%26st.mrkId%3D%257B%2522uploadedMovieMarker%2522%253A%257B%2522marker%2522%253A%25221519410114503%2522%252C%2522hasMore%2522%253Atrue%257D%252C%2522sharedMovieMarker%2522%253A%257B%2522marker%2522%253Anull%252C%2522hasMore%2522%253Afalse%257D%257D%26st.friendId%3D561722190321%26st.frwd%3Don%26_prevCmd%3DfriendMovies%26tkn%3D7257&st.discType=MOVIE&st.mvId=863789452017&_prevCmd=friendMovies&tkn=3648#lst#',
 206         'only_matching': True,
 207     }, {
 208         # Paid video
 209         'url': 'https://ok.ru/video/954886983203',
 210         'only_matching': True,
 211     }, {
 212         'url': 'https://ok.ru/videoembed/2932705602075',
 213         'info_dict': {
 214             'id': '2932705602075',
 215             'ext': 'mp4',
 216             'thumbnail': 'https://i.mycdn.me/videoPreview?id=1369902483995&type=37&idx=2&tkn=fqlnoQD_xwq5ovIlKfgNyU08qmM&fn=external_8',
 217             'title': 'Boosty для тебя!',
 218             'uploader_id': '597811038747',
 219             'like_count': 0,
 220             'duration': 35,
 221         },
 222     }]
 223
 224     _WEBPAGE_TESTS = [{
 225         'url': 'https://boosty.to/ikakprosto/posts/56cedaca-b56a-4dfd-b3ed-98c79cfa0167',
 226         'info_dict': {
 227             'id': '3950343629563',
 228             'ext': 'mp4',
 229             'thumbnail': 'https://i.mycdn.me/videoPreview?id=2776238394107&type=37&idx=11&tkn=F3ejkUFcpuI4DnMRxrDGcH5YcmM&fn=external_8',
 230             'title': 'Заяц Бусти.mp4',
 231             'uploader_id': '571368965883',
 232             'like_count': 0,
 233             'duration': 10444,
 234         },
 235         'skip': 'Site no longer embeds',
 236     }]
 237
 238     def _clear_cookies(self, cdn_url):
 239         # Direct http downloads will fail if CDN cookies are set
 240         # so we need to reset them after each format extraction
 241         if self._get_cookies('https://notarealsubdomain.mycdn.me/'):
 242             self.cookiejar.clear(domain='.mycdn.me')
 243         if self._get_cookies(cdn_url):
 244             self.cookiejar.clear(domain=urllib.parse.urlparse(cdn_url).hostname)
 245
 246     @classmethod
 247     def _extract_embed_urls(cls, url, webpage):
 248         for x in super()._extract_embed_urls(url, webpage):
 249             yield smuggle_url(x, {'referrer': url})
 250
 251     def _real_extract(self, url):
 252         try:
 253             return self._extract_desktop(url)
 254         except ExtractorError as e:
 255             try:
 256                 return self._extract_mobile(url)
 257             except ExtractorError:
 258                 # error message of desktop webpage is in English
 259                 raise e
 260
 261     def _extract_desktop(self, url):
 262         start_time = int_or_none(compat_parse_qs(
 263             compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
 264
 265         url, smuggled = unsmuggle_url(url, {})
 266         video_id, is_embed = self._match_valid_url(url).group('id', 'embed')
 267         mode = 'videoembed' if is_embed else 'video'
 268
 269         webpage = self._download_webpage(
 270             f'https://ok.ru/{mode}/{video_id}', video_id,
 271             note='Downloading desktop webpage',
 272             headers={'Referer': smuggled['referrer']} if smuggled.get('referrer') else {})
 273
 274         error = self._search_regex(
 275             r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<',
 276             webpage, 'error', default=None)
 277         # Direct link from boosty
 278         if (error == 'The author of this video has not been found or is blocked'
 279                 and not smuggled.get('referrer') and mode == 'videoembed'):
 280             return self._extract_desktop(smuggle_url(url, {'referrer': 'https://boosty.to'}))
 281         elif error:
 282             raise ExtractorError(error, expected=True)
 283
 284         player = self._parse_json(
 285             unescapeHTML(self._search_regex(
 286                 r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
 287                 webpage, 'player', group='player')),
 288             video_id)
 289
 290         # embedded external player
 291         if player.get('isExternalPlayer') and player.get('url'):
 292             return self.url_result(player['url'])
 293
 294         flashvars = player['flashvars']
 295
 296         metadata = flashvars.get('metadata')
 297         if metadata:
 298             metadata = self._parse_json(metadata, video_id)
 299         else:
 300             data = {}
 301             st_location = flashvars.get('location')
 302             if st_location:
 303                 data['st.location'] = st_location
 304             metadata = self._download_json(
 305                 compat_urllib_parse_unquote(flashvars['metadataUrl']),
 306                 video_id, 'Downloading metadata JSON',
 307                 data=urlencode_postdata(data))
 308
 309         movie = metadata['movie']
 310
 311         # Some embedded videos may not contain title in movie dict (e.g.
 312         # http://ok.ru/video/62036049272859-0) thus we allow missing title
 313         # here and it's going to be extracted later by an extractor that
 314         # will process the actual embed.
 315         provider = metadata.get('provider')
 316         title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
 317
 318         thumbnail = movie.get('poster')
 319         duration = int_or_none(movie.get('duration'))
 320
 321         author = metadata.get('author', {})
 322         uploader_id = author.get('id')
 323         uploader = author.get('name')
 324
 325         upload_date = unified_strdate(self._html_search_meta(
 326             'ya:ovs:upload_date', webpage, 'upload date', default=None))
 327
 328         age_limit = None
 329         adult = self._html_search_meta(
 330             'ya:ovs:adult', webpage, 'age limit', default=None)
 331         if adult:
 332             age_limit = 18 if adult == 'true' else 0
 333
 334         like_count = int_or_none(metadata.get('likeCount'))
 335
 336         subtitles = {}
 337         for sub in traverse_obj(metadata, ('movie', 'subtitleTracks', ...), expected_type=dict):
 338             sub_url = sub.get('url')
 339             if not sub_url:
 340                 continue
 341             subtitles.setdefault(sub.get('language') or 'en', []).append({
 342                 'url': sub_url,
 343                 'ext': 'vtt',
 344             })
 345
 346         info = {
 347             'id': video_id,
 348             'title': title,
 349             'thumbnail': thumbnail,
 350             'duration': duration,
 351             'upload_date': upload_date,
 352             'uploader': uploader,
 353             'uploader_id': uploader_id,
 354             'like_count': like_count,
 355             'age_limit': age_limit,
 356             'start_time': start_time,
 357             'subtitles': subtitles,
 358         }
 359
 360         # pladform
 361         if provider == 'OPEN_GRAPH':
 362             info.update({
 363                 '_type': 'url_transparent',
 364                 'url': movie['contentId'],
 365             })
 366             return info
 367
 368         if provider == 'USER_YOUTUBE':
 369             info.update({
 370                 '_type': 'url_transparent',
 371                 'url': movie['contentId'],
 372             })
 373             return info
 374
 375         assert title
 376         if provider == 'LIVE_TV_APP':
 377             info['title'] = title
 378
 379         quality = qualities(('4', '0', '1', '2', '3', '5', '6', '7'))
 380
 381         formats = [{
 382             'url': f['url'],
 383             'ext': 'mp4',
 384             'format_id': f.get('name'),
 385         } for f in traverse_obj(metadata, ('videos', lambda _, v: url_or_none(v['url'])))]
 386
 387         m3u8_url = traverse_obj(metadata, 'hlsManifestUrl', 'ondemandHls')
 388         if m3u8_url:
 389             formats.extend(self._extract_m3u8_formats(
 390                 m3u8_url, video_id, 'mp4', 'm3u8_native',
 391                 m3u8_id='hls', fatal=False))
 392             self._clear_cookies(m3u8_url)
 393
 394         for mpd_id, mpd_key in [('dash', 'ondemandDash'), ('webm', 'metadataWebmUrl')]:
 395             mpd_url = metadata.get(mpd_key)
 396             if mpd_url:
 397                 formats.extend(self._extract_mpd_formats(
 398                     mpd_url, video_id, mpd_id=mpd_id, fatal=False))
 399                 self._clear_cookies(mpd_url)
 400
 401         dash_manifest = metadata.get('metadataEmbedded')
 402         if dash_manifest:
 403             formats.extend(self._parse_mpd_formats(
 404                 compat_etree_fromstring(dash_manifest), 'mpd'))
 405
 406         for fmt in formats:
 407             fmt_type = self._search_regex(
 408                 r'\btype[/=](\d)', fmt['url'],
 409                 'format type', default=None)
 410             if fmt_type:
 411                 fmt['quality'] = quality(fmt_type)
 412
 413         # Live formats
 414         m3u8_url = metadata.get('hlsMasterPlaylistUrl')
 415         if m3u8_url:
 416             formats.extend(self._extract_m3u8_formats(
 417                 m3u8_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 418             self._clear_cookies(m3u8_url)
 419         rtmp_url = metadata.get('rtmpUrl')
 420         if rtmp_url:
 421             formats.append({
 422                 'url': rtmp_url,
 423                 'format_id': 'rtmp',
 424                 'ext': 'flv',
 425             })
 426
 427         if not formats:
 428             payment_info = metadata.get('paymentInfo')
 429             if payment_info:
 430                 self.raise_no_formats('This video is paid, subscribe to download it', expected=True)
 431
 432         info['formats'] = formats
 433         return info
 434
 435     def _extract_mobile(self, url):
 436         video_id = self._match_id(url)
 437
 438         webpage = self._download_webpage(
 439             'http://m.ok.ru/video/%s' % video_id, video_id,
 440             note='Downloading mobile webpage')
 441
 442         error = self._search_regex(
 443             r'видео</a>\s*<div\s+class="empty">(.+?)</div>',
 444             webpage, 'error', default=None)
 445         if error:
 446             raise ExtractorError(error, expected=True)
 447
 448         json_data = self._search_regex(
 449             r'data-video="(.+?)"', webpage, 'json data')
 450         json_data = self._parse_json(unescapeHTML(json_data), video_id) or {}
 451
 452         redirect_url = self._request_webpage(HEADRequest(
 453             json_data['videoSrc']), video_id, 'Requesting download URL').geturl()
 454         self._clear_cookies(redirect_url)
 455
 456         return {
 457             'id': video_id,
 458             'title': json_data.get('videoName'),
 459             'duration': float_or_none(json_data.get('videoDuration'), scale=1000),
 460             'thumbnail': json_data.get('videoPosterSrc'),
 461             'formats': [{
 462                 'format_id': 'mobile',
 463                 'url': redirect_url,
 464                 'ext': 'mp4',
 465             }]
 466         }